Pasquale Minervini
- Papers
- 23
Cite
Notes
Only stored in your browser.
Authored papers
23MMLongBench: Benchmarking Long-Context Vision-Language Models Effectively and Thoroughly
arXiv 2025
Neurosymbolic Diffusion Models
arXiv 2025
Q-Filters: Leveraging QK Geometry for Efficient KV Cache Compression
arXiv 2025
OpenSIR: Open-Ended Self-Improving Reasoner
arXiv 2025
PosterSum: A Multimodal Benchmark for Scientific Poster Summarization
arXiv 2025
Self-Training Large Language Models for Tool-Use Without Demonstrations
arXiv 2025
Inverse Scaling in Test-Time Compute
arXiv 2025
Large language models surpass human experts in predicting neuroscience results
arXiv 2024
Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering
arXiv 2024
Analysing the Residual Stream of Language Models Under Knowledge Conflicts
arXiv 2024
Are We Done with MMLU?
arXiv 2024
DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations
arXiv 2024
Analysing The Impact of Sequence Composition on Language Model Pre-Training
arXiv 2024
A Simple and Effective $L_2$ Norm-Based Strategy for KV Cache Compression
arXiv 2024
No Train No Gain: Revisiting Efficient Training Algorithms For Transformer-based Language Models
NeurIPS 2023 11
Parameter-Efficient Fine-Tuning of LLaMA for the Clinical Domain
arXiv 2023
Adaptive Computation Modules: Granular Conditional Computation For Efficient Inference
arXiv 2023
SPARSEFIT: Few-shot Prompting with Sparse Fine-tuning for Jointly Generating Predictions and Natural Language Explanations
arXiv 2023
Using Natural Language Explanations to Improve Robustness of In-context Learning
arXiv 2023
An Efficient Memory-Augmented Transformer for Knowledge-Intensive NLP Tasks
arXiv 2022
MedDistant19: Towards an Accurate Benchmark for Broad-Coverage Biomedical Relation Extraction
COLING 2022 10
XQA-DST: Multi-Domain and Multi-Lingual Dialogue State Tracking
arXiv 2022
PAQ: 65 Million Probably-Asked Questions and What You Can Do With Them
arXiv 2021
Affiliations
Frequent co-authors
10from 23 papers